CN103616036B - A kind of airborne sensor systematic error estimation based on cooperative target and compensation method - Google Patents

A kind of airborne sensor systematic error estimation based on cooperative target and compensation method Download PDF

Info

Publication number
CN103616036B
CN103616036B CN201310630077.2A CN201310630077A CN103616036B CN 103616036 B CN103616036 B CN 103616036B CN 201310630077 A CN201310630077 A CN 201310630077A CN 103616036 B CN103616036 B CN 103616036B
Authority
CN
China
Prior art keywords
target
sensor
deviation
coordinate
compensation
Prior art date
Application number
CN201310630077.2A
Other languages
Chinese (zh)
Other versions
CN103616036A (en
Inventor
徐飞
黄大羽
Original Assignee
中国航空无线电电子研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国航空无线电电子研究所 filed Critical 中国航空无线电电子研究所
Priority to CN201310630077.2A priority Critical patent/CN103616036B/en
Publication of CN103616036A publication Critical patent/CN103616036A/en
Application granted granted Critical
Publication of CN103616036B publication Critical patent/CN103616036B/en

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/40Means for monitoring or calibrating

Abstract

The present invention proposes a kind of airborne sensor systematic error estimation based on cooperative target and compensation method.The present invention is the observation data configuration systematic error equation to it first with cooperative target self poisoning information and airborne sensor;Then use equivalent method, use auto-adaptive filtering technique estimate measurement system deviation and determine the systematic error that appearance deviation is superimposed;Finally according to search coverage, noncooperative target metric data is carried out partition compensation.Experiment proves that strong robustness of the present invention, reliability are high, computation complexity is low, is particularly well-suited to engineering practice.

Description

A kind of airborne sensor systematic error estimation based on cooperative target and compensation method

Technical field

Inventive algorithm relates to Data fusion technique, particularly to sensor space registration technique, Specifically refer to a kind of airborne sensor systematic error estimation based on cooperative target and compensation method.

Background technology

Modern war is faced with complication, diversified external environment condition, and single-sensor is difficult to Meet the demand that battle field information is analyzed.Increasing it was verified that use multisensor syste Select suitable algorithm that multi-source information is carried out fusion treatment, it is possible to obtain than single-sensor more The useful information of horn of plenty.Compared to Method for Single Sensor System, multisensor syste enhances system Survival ability, extend time covering domain and space covering domain, improve the credibility of information. But, vital premise is that these advantages produce accurately solves to go out in fusion process Existing a series of key issues.Spatial registration is one of them key issue, and complete space is joined Quasi-Resolving probiems comprises the intension of two aspects: registration error estimation and system deviation compensate.Mesh Front space registration problems mainstream research concentrates on the former, and the research to the latter is then mainly reflected in work Journey should be considered by the actual of used time.

Mainly have than more typical systematic error estimation method: real-time quality control methods, a young waiter in a wineshop or an inn Multiplication, maximum likelihood method, Kalman filtering method etc..Real-time quality control methods is to each sensor Measured metric data is averaging processing, and the observation as sensor of averaging is entered And estimating system deviation.This method ignores sensor measurement noise and each sensor relative to public affairs The deviation of the coordinate system impact on registration error altogether, it is adaptable to the situation that measurement noise is less.? Registration problems is converted into least-squares parameter estimation problem by little square law, by deviation equation Structure and solving of over-determined systems obtain straggling parameter estimation under least square meaning.This The method of kind is applicable to the registration error estimation of offline mode, because the structure of over-determined systems needs The sensor metric data in multiple moment.Maximum likelihood method utilizes sensor in System planes Measured value, uses maximum likelihood method to estimate position and the system deviation of target simultaneously, It has used two step recursive optimization methods to accelerate the convergence rate estimated.Kalman filter method is recognized For constant and unrelated with noise during system deviation vector, by structural regime equation and measurement side Journey realizes the estimation of system deviation.Certainly, researcher is also developed other system deviation and is estimated Meter method, or the impact improving or adding some factor of said method.But these The basic ideas of method are consistent: first instrument error equation;Secondly analytical error source, Linearized stability equation;Last according to Parameter Estimation Problem solving system deviation.

Make a general survey of above-mentioned canonical system error estimation it appeared that But most of algorithms can only measure Data random error is less even without just effective during random error, this more exacting terms It is implacable in practice.Have ignored partially it addition, more important point is said method The consideration of difference estimation observability, the especially registration error estimation under mobile platform, now Deviation variables not only contains sensor Measurement Biases and also includes platform self navigation deviation, system Observability problem become especially prominent.

System deviation compensates and is primarily referred to as the apriority of deviation profile in sensor detection spatial domain Assuming that.One is to think that deviation is invariable, is evenly distributed in detection spatial domain;Another kind is recognized Slowly varying for deviation, in detection spatial domain non-uniform Distribution.More realistic is that latter is false If systematic error is the multi-variable function with the change of detection spatial domain, by partially to have researcher to assume Difference estimated value utilizes least square fitting to go out function coefficients.This method needs substantial amounts of number According to, and, matching order and Variable selection are all suitable stubborn problems.

Summary of the invention

The defect existed with compensation technique for existing sensing system error estimation, the present invention carries Go out a kind of airborne sensor systematic error estimation based on cooperative target and compensation method, the present invention The observation data configuration to it first with cooperative target self poisoning information and airborne sensor Systematic error equation;Then use equivalent method, use auto-adaptive filtering technique to estimate measurement System deviation and determine the systematic error that appearance deviation is superimposed;Finally according to search coverage to non- Cooperative target metric data carries out partition compensation.

The goal of the invention of the present invention is achieved through the following technical solutions:

The first step, by distinguishing inspection cooperative target and noncooperative target data;Specific practice is

It is divided into N number of region, each region to set two and deposit sensor space exploration by rule Storage area: cooperative target district and noncooperative target district.According to target metric data and target whether It is that cooperative target stores data into respective regions.

Second step, utilizes the location data of cooperative target and sensor to its metric data instrument error side Journey;Specific practice is

(1) cooperative target location data are transformed into ECEF coordinate by geodetic coordinates

The geodetic coordinates of target location is (L, λ, H), and wherein, L represents latitude, and λ is longitude, H is height, then its corresponding ECEF coordinate (xe,ye,ze) it is

x e = ( C + H ) cos L c o s λ y e = ( C + H ) cos L s i n λ z e = [ C ( 1 - e 2 ) + H ] sin L ,

Wherein

C = E q ( 1 - e 2 sin 2 L ) 1 / 2 ,

EqBeing equatorial radius, e is eccentricity of the earth.

(2) by ECEF Coordinate Conversion to local rectangular coordinates

Local rectangular coordinate system usually approximates inertial coodinate system.ECEF coordinate is to partial, right angle The conversion of coordinate is typically through rotating and two links of translation, and wherein spin matrix is generally according to seat Parameter to definition and direction cosines solve.

Target is [x at ECEF coordinatee,ye,ze]T, carrier aircraft ECEF coordinate now is [xeo,yeo,zeo]T, it is C that ECEF coordinate is tied to local rectangular coordinate system spin matrix1, then mesh The coordinate being marked on local rectangular coordinate system is

x g y g z g = C 1 ( x e y e z e - x e o y e o z e o ) ,

(3) it is transformed into sensor measurement coordinate by local rectangular coordinates

Sensor measurement coordinate system is typically a kind of non-stable coordinate system, and it is sat with partial, right angle Conversion between mark system generally need to use the attitude angle information of sensor (when being rigidly connected, also referred to as For platform stance angle, for yaw angle α, pitching angle beta, roll angle γ).

The local rectangular coordinates of target is [xg,yg,zg]T, local rectangular coordinate system is to sensor The spin matrix measuring coordinate system (rectangular system) is C2(C2Value is determined by platform stance angle), Then target at the coordinate of sensor measurement coordinate system (rectangular system) is

x b y b z b = C 2 x g y g z g ,

Target is at the coordinate distance ρ of sensor measurement coordinate system (ball system)d, azimuth Pitching angle thetad, for

ρ d = x b 2 + y b 2 + z b 2 ,

θ d = tan - 1 ( z b x b 2 + y b 2 ) ,

(4) cooperative target measurement information instrument error equation is combined

Sensor is distance ρ to the metric data of targetm=ρ '+Δ ρ+vρ(t), orientation AnglePitching angle thetam=θ '+Δ θ+vθ(t);Wherein, Respectively measure true value,For Measurement Biases,For random error. Platform stance angle α=α '+Δ α, β=β '+Δ beta, gamma=γ '+Δ γ, α ', β ', γ ' is respectively corresponding true value, Δ α, Δ β, Δ γ is for determine appearance deviation.Error between target metric data and object location data is Δ α, Δ β, Δ γ,Function, then error equation is represented by

Wherein,

3rd step, analysis deviation source, select suitable bias vector, linearized stability equation;Specifically Way is

(1) deviation variables is selected

ξmdBeing the function about six deviation variables, traditional way is through complicated Derivation is tried to achieveAnalytic expression, then f () is carried out one Rank Taylor expansion can obtain the approximately linear expression formula of f ().It will be clear that now need to estimate Parameter more, only with ξmdThree-dimensional information is difficult to ensure that the observability of system.Therefore, Select hereinIt is added to as Δ α, Δ β, Δ γAfter on Equivalence Measurement Biases as system deviation vector.

(2) linearisation

Select equivalence Measurement BiasesAs system deviation, then linearisation just becomes non- The simplest, for

Wherein, H = 1 0 0 0 1 0 0 0 1 .

4th step, affects the different occasions of filtering performance according to noise characteristic, improves adaptive-filtering skill Art presses subregion estimated bias parameter;Specific practice is

(1) structural regime equation and measurement equation

OrderAssume that equivalent deviation is slowly varying, then state equation is

Xk+1=Xk+wk,

Measurement equation is

Z=HXk+1+vk,

(2) filtering initializes

Estimate initial valueInitial estimation error battle arrayMeasurement noise initial variance Battle arrayProcess noise initial variance battle arrayAdaptive-filtering attenuation quotient &beta; j = 1 - &lambda; 1 - &lambda; k + 1 &CenterDot; &lambda; k - j , 0 < &lambda; < 1.

(3) time updates and process-noise variance battle array ART network

One-step prediction:

X ^ k | k - 1 = F k - 1 X ^ k - 1 | k - 1 ,

Predicting covariance battle array:

P k | k - 1 = F k - 1 P k - 1 | k - 1 F k - 1 T + Q ^ k - 1 ,

Measuring noise square difference battle array ART network

R ^ k = &beta; k Z ~ k Z ~ k T + ( 1 - &beta; k ) R ^ k - 1 ,

(4) renewal and process-noise variance battle array ART network are measured

Filtering gain battle array:

K k = P k | k - 1 H k T ( H k P k | k - 1 H k T + R ^ k ) - 1 ,

Estimation updated value:

X ^ k | k = X ^ k | k - 1 + K k ( Z k - H k X ^ k | k - 1 ) ,

Estimation difference covariance matrix:

Pk|k=(I-KkHk)Pk|k-1,

Process-noise variance battle array ART network:

Q ^ k = &beta; k K k Z ~ k Z ~ k T K k T + ( 1 - &beta; k ) Q ^ k - 1 ,

5th step, carries out partition compensation to noncooperative target metric data, obtains the amount after deviation compensation Survey data;Specific practice is

(1) subregion belonging to noncooperative target is defined

The principle that subregion belonging to noncooperative target defines is that the target information detected according to sensor is Regionalization basis.During compensation, each bat all needs to calculate.

(2) compensation is measured

Noncooperative target measuresThe n-th etc. of subregion belonging to this target Effect estimation of deviation result isMetric data after then compensating is

&zeta; &OverBar; m = &zeta; m X ^ n ,

Measurement used by subsequent treatment (such as filtering) just usesζ m.It addition, when target is from one point When district enters another subregion, the corresponding equivalent deviation compensated also to change therewith.

Compared with prior art, the present invention first avoids multiple error parametric variable and estimates simultaneously Time the most considerable problem of system;Second, solve system by mistake by introducing auto-adaptive filtering technique Slowly varying and other unknown noises of difference exist in the case of estimation problem;3rd, to non-conjunction Making target measurement takes partition compensation to compensate residual error with further reduction according to region.Experiment proves Strong robustness of the present invention, reliability are high, computation complexity is low, are particularly well-suited to engineering practice.

Accompanying drawing explanation

Fig. 1 is fundamental diagram of the present invention

Fig. 2 is the 1st subregion equivalent distances estimation of deviation result

Fig. 3 is the 1st subregion equivalence azimuth deviation estimated result

Fig. 4 is the 1st subregion equivalence pitch deviation estimated result

Fig. 5 is noncooperative target distance measuring deviation compensation whether results contrast

Fig. 6 is noncooperative target azimuthal measuring deviation compensation whether results contrast

Fig. 7 is whether noncooperative target pitching Measurement Biases compensates results contrast

Fig. 8 be noncooperative target measure whether compensate filter after RMS compare

Detailed description of the invention

The invention will be further described below in conjunction with the accompanying drawings: the present embodiment is with the technology of the present invention Implement under premised on scheme, give detailed embodiment and concrete operating process, But protection scope of the present invention is not limited to following embodiment.

This section illustrates for embodiment with onboard radar system error estimation with compensating, the present embodiment bag Include following steps:

The first step, by distinguishing inspection cooperative target and noncooperative target data;It is specially

It is divided into N number of region, each region to set two sensor space exploration by certain rule Individual memory block: cooperative target district and noncooperative target district.According to target metric data and target Whether it is that cooperative target stores data into respective regions.

Second step, utilizes the location data of cooperative target and sensor to its metric data instrument error side Journey;Concretely comprise the following steps

Local rectangular coordinate system (carrier aircraft geographic coordinate system) selects sky, northeast system.Sensor measurement Coordinate system is consistent with carrier aircraft coordinate system is spherical coordinate system, and carrier aircraft coordinate system is former with carrier aircraft barycenter Point, x-axis along carrier transverse axis to the right, y-axis along before carrier Y-direction, z-axis along carrier vertical pivot to On.

(1) cooperative target location data are transformed into ECEF coordinate by geodetic coordinates

The geodetic coordinates of target location is (L, λ, H), and wherein, L represents latitude, and λ is longitude, H is height, then its corresponding ECEF coordinate (xe,ye,ze) it is

x e = ( C + H ) cos L c o s &lambda; y e = ( C + H ) cos L s i n &lambda; z e = &lsqb; C ( 1 - e 2 ) + H &rsqb; sin L ,

Wherein

C = E q ( 1 - e 2 sin 2 L ) 1 / 2 ,

EqBeing equatorial radius, e is eccentricity of the earth.

(2) by ECEF Coordinate Conversion to carrier aircraft geographical coordinate

ECEF coordinate system is transformed into the spin matrix of carrier aircraft geographic coordinate system (sky, northeast system) (λ, L are respectively carrier aircraft place longitude, latitude)

C 1 = - s i n &lambda; c o s &lambda; 0 - sin L cos &lambda; - sin L s i n &lambda; cos L cos L c o s &lambda; cos L sin &lambda; sin L ,

Target is [x at ECEF coordinatee,ye,ze]T, carrier aircraft ECEF coordinate now is [xeo,yeo,zeo]T, it is C that ECEF coordinate is tied to carrier aircraft geographic coordinate system spin matrix1, then target Coordinate in carrier aircraft geographic coordinate system is

x g y g z g = C 1 ( x e y e z e - x e o y e o z e o ) ,

(3) it is transformed into carrier aircraft coordinate by carrier aircraft geographical coordinate

Carrier aircraft geographical coordinate is tied to the spin matrix of carrier aircraft coordinate system for (α, β, γ are respectively carrier aircraft Yaw angle, the angle of pitch, roll angle)

C 2 = cos &gamma; cos &alpha; - sin &gamma; sin &beta; sin &alpha; cos &gamma; sin &alpha; + sin &gamma; sin &beta; cos &alpha; - sin &gamma; cos &beta; - cos &beta; sin &alpha; cos &beta; cos &alpha; sin &beta; sin &gamma; cos &alpha; + cos &gamma; sin &beta; sin &alpha; sin &gamma; sin &alpha; - cos &gamma; sin &beta; cos &alpha; cos &gamma; cos &beta; ,

The carrier aircraft geographical coordinate of target is [xg,yg,zg]T, then target is at carrier aircraft coordinate system (right angle System) coordinate be

x b y b z b = C 2 x g y g z g ,

It is transformed under carrier aircraft spherical coordinate system, then distance ρd, azimuthPitching angle thetadIt is respectively

&rho; d = x b 2 + y b 2 + z b 2 ,

&theta; d = tan - 1 ( z b x b 2 + y b 2 ) ,

(4) instrument error equation

Wherein,Represent respectively target metric data and Data under the measurement coordinate system come by targeting information conversion; Δ θ is respectively needs (platform) yaw angle deviation of estimation, (platform) pitch angle deviation, horizontal stroke Roll angle deviation, (target) measure range deviation, (target) measures azimuth deviation, (aim parameter Survey) pitch deviation.

3rd step, analysis deviation source, select suitable bias vector, linearized stability equation;Specifically Step is

SelectEstimate as equivalent distance, orientation, pitch deviation, linearisation Error equation be

4th step, auto-adaptive filtering technique based on a kind of improvement presses subregion estimated bias parameter;Specifically Step is

Use adaptive filter algorithm to carry out estimation of deviation, state equation and measurement equation to be respectively

Xk+1=Xk+wk,

Z=HXk+1+vk,

Estimate initial valueInitial estimation error battle arrayMeasure Noise initial variance battle arrayProcess noise initial variance battle arrayAdaptive-filtering declines Subtract coefficientWherein λ=0.2.Then estimation of deviation based on adaptive-filtering is pressed Equation below is carried out.

X ^ k | k - 1 = F k - 1 X ^ k - 1 | k - 1 P k | k - 1 = F k - 1 P k - 1 | k - 1 F k - 1 T + G ^ k - 1 S ^ k = &beta; k Z ~ k Z ~ k T + ( 1 - &beta; k ) S ^ k - 1 K k = P k | k - 1 H k T ( H k P k | k - 1 H k T + S ^ k ) - 1 X ^ k | k = X ^ k | k - 1 + K k ( Z k - H k X ^ k | k - 1 ) P k | k = ( I - K k H k ) P k | k - 1 G ^ k = &beta; k K k Z ~ k Z ~ k T K k T + ( 1 - &beta; k ) G ^ k - 1 ,

5th step, carries out partition compensation to noncooperative target metric data, obtains the amount after deviation compensation Survey data;Concretely comprise the following steps

First subregion belonging to current bat time noncooperative target measurement is calculated, then from straggling parameter storehouse The corresponding parameter value of middle extraction compensates.Noncooperative target measures The n-th equivalent deviation estimated result of this target now affiliated subregion isAmount after then compensating Survey data are

&zeta; &OverBar; m = &zeta; m - X ^ n ,

Measurement used by subsequent treatment (such as filtering) just usesζ m

Test case

Arrange radar measurement distance random error standard deviation be 100 meters, system of distance deviation be + 30 meters, radar measurement orientation random error standard deviation is 0.5 degree, azimuth system deviation is+1 Degree, radar measurement pitching random error standard deviation is 0.5 degree, pitch system deviation is+0.5 degree. Carrier aircraft platform navigation is determined appearance deviation (yaw angle, the angle of pitch, roll angle) and is+0.1 degree. Cooperative target (quantity is some) be evenly distributed in sensor detection areas, noncooperative target (number Measure 1) with the distance of carrier aircraft at 80 km.8 districts are divided in sensor detection areas.

Fig. 2 to Fig. 4 is the 1st subregion equivalent distances deviation, equivalence azimuth deviation and equivalence respectively Pitch deviation estimated result.It is found that because the existence of carrier aircraft platform navigation attitude angle deviation, One is had between equivalent deviation (orientation, pitching) and radar measurement deviation (azimuth pitch) The difference of relative constancy.Other subregions can be similar to experimental result.

Fig. 5 to Fig. 7 be respectively noncooperative target after systematic error partition compensation with compensate before Metric data contrast.Because the existence of systematic error, the metric data before compensation and true value it Between there is the droop of relative constancy;And after partition compensation, this relatively-stationary partially Difference reduces (contrast on pitch channel is especially apparent).

Fig. 8 is that noncooperative target measures the contrast whether compensating filter result, and both difference are The most obvious.This is because target metric data is not only polluted by sensor Measurement Biases Also receive the pollution of navigation attitude misalignment, if both impacts, target can not be effectively eliminated Following the tracks of result will be the most undesirable.The present invention can successfully solve two kinds of simultaneous feelings of deviation Condition, therefore after metric data is compensated, tracking performance is greatly improved.

It is above the present invention preferably embodiment, but those skilled in the art should manage Solving, these are merely illustrative of, on the premise of without departing substantially from the principle of the present invention and essence, and can So that these embodiments are made various changes or modifications.Therefore, protection scope of the present invention by Appended claims limits.

Claims (5)

1. airborne sensor systematic error estimation based on cooperative target and a compensation method, It is characterized in that comprising the following steps:
The first step, by distinguishing inspection cooperative target and noncooperative target data;
Second step, utilizes the location data of cooperative target and sensor to miss its metric data structure Eikonal equation;
3rd step, analysis deviation source, select suitable bias vector, linearized stability equation;
4th step, affects the different occasions of filtering performance according to noise characteristic, improves self adaptation filter Wave technology presses subregion estimated bias parameter;
5th step, carries out partition compensation to noncooperative target metric data, after obtaining deviation compensation Metric data.
Airborne sensor systematic error estimation the most according to claim 1 and compensation method, It is characterized in that: the instrument error equation described in second step, comprise the steps of
21), cooperative target positions data by geodetic coordinates converting into target at ECEF coordinate (xe,ye,ze) it is:
x e = ( C + H ) cos L c o s &lambda; y e = ( C + H ) cos L s i n &lambda; z e = &lsqb; C ( 1 - e 2 ) + H &rsqb; sin L ,
Wherein, L represents latitude, and λ is longitude, and H is height,EqIt it is equator Radius, e is eccentricity of the earth;
22), cooperative target by ECEF Coordinate Conversion to target at local rectangular coordinates is:
x g y g z g = C 1 ( x e y e z e - x e o y e o z e o ) ,
Wherein, [xeo,yeo,zeo]TFor carrier aircraft ECEF coordinate, C1It is tied to local straight for ECEF coordinate Angle coordinate system spin matrix;
23), change at local rectangular coordinates by target according to the platform stance angle information of sensor To target at sensor measurement coordinate it is:
x b y b z b = C 2 x g y g z g ,
Wherein, C2For the spin matrix of local rectangular coordinate system to sensor measurement coordinate, C2Value by Platform stance angle determines, target is at the coordinate distance ρ of sensor measurement coordinate systemd, azimuthPitching angle thetad, for:
&rho; d = x b 2 + y b 2 + z b 2 ,
&theta; d = tan - 1 ( z b x b 2 + y b 2 ) ;
24) cooperative target location information, is combined with sensor to its metric data equationof structure:
Wherein, ξmFor sensor metric data,Distance ρm=ρ '+Δ ρ+vρ(t), azimuthPitching angle thetam=θ '+Δ θ+vθ(t), Wherein, ρ ',θ ' respectively measures true value, Δ ρ,Δ θ is Measurement Biases, vρ(t),vθ(t) For random error;ξdIt is the targeting information numerical value that is transformed under sensor measurement coordinate system,Δα,Δβ,Δγ,Δρ,Δ θ representative sensor carrying platform driftage respectively Angular displacement, carrying platform pitch angle deviation, carrying platform roll angle deviation, sensor distance amount Survey deviation, sensor orientation Measurement Biases, sensor pitching Measurement Biases.
Airborne sensor systematic error estimation the most according to claim 2 and compensation method, It is characterized in that: the bias vector described in the 3rd step is Δ α, Δ β, Δ γ is added to Δ ρ,Δθ Form the bias vector of equivalence afterwardsDescribed linearized stability equation, for
Wherein,
Airborne sensor systematic error estimation the most according to claim 3 and compensation method, It is characterized in that: the estimated bias parameter described in the 4th step comprises the steps of
41), the structure state equation of estimation of deviation and measurement equation, for
OrderAssume that equivalent deviation is slowly varying, then state equation:
Xk+1=Xk+wk,
Measurement equation:
Z=HXk+1+vk,
Wherein, wkIt it is process noise;vkIt it is measurement noise;
42), on the basis of qualitative analysis, the difference of filtering performance is affected according to noise characteristic Occasion, indirectly makeover process noise and measurement noise characteristic, detailed process is as follows:
(1) filtering initializes:
Estimate initial valueInitial estimation error battle arrayMeasurement noise initial variance Battle arrayProcess noise initial variance battle arrayAdaptive-filtering attenuation quotient0 < λ < 1;
(2) time updates and process-noise variance battle array ART network:
One-step prediction:
X ^ k | k - 1 = F k - 1 X ^ k - 1 | k - 1 ,
Predicting covariance battle array:
P k | k - 1 = F k - 1 P k - 1 | k - 1 F k - 1 T + Q ^ k - 1 ,
Measuring noise square difference battle array ART network:
R ^ k = &beta; k Z ~ k Z ~ k T + ( 1 - &beta; k ) R ^ k - 1 ,
(3) renewal and process-noise variance battle array ART network are measured:
Filtering gain battle array:
K k = P k | k - 1 H k T ( H k P k | k - 1 H k T + R ^ k ) - 1 ,
Estimation updated value:
X ^ k | k = X ^ k | k - 1 + K k ( Z k - H k X ^ k | k - 1 ) ,
Estimation difference covariance matrix:
Pk|k=(I-KkHk)Pk|k-1,
Process-noise variance battle array ART network:
Q ^ k = &beta; k K k Z ~ k Z ~ k T K k T + ( 1 - &beta; k ) Q ^ k - 1 .
Airborne sensor systematic error estimation the most according to claim 1 and compensation method, It is characterized in that: described in the 5th step, noncooperative target metric data is carried out partition compensation, Comprise the steps of
It is first according to noncooperative target metric data and defines subregion belonging to it;Then from corresponding subregion In straggling parameter storehouse, extracting parameter value complement is repaid, and its compensation formula is:
&zeta; &OverBar; m = &zeta; m - X ^ n ,
Wherein,The n-th equivalent deviation estimated result of subregion, ζ belonging to this targetmFor non-conjunction Make target to measure,ζ mFor the metric data after compensating.
CN201310630077.2A 2013-11-29 2013-11-29 A kind of airborne sensor systematic error estimation based on cooperative target and compensation method CN103616036B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201310630077.2A CN103616036B (en) 2013-11-29 2013-11-29 A kind of airborne sensor systematic error estimation based on cooperative target and compensation method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201310630077.2A CN103616036B (en) 2013-11-29 2013-11-29 A kind of airborne sensor systematic error estimation based on cooperative target and compensation method

Publications (2)

Publication Number Publication Date
CN103616036A CN103616036A (en) 2014-03-05
CN103616036B true CN103616036B (en) 2016-12-07

Family

ID=50166742

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201310630077.2A CN103616036B (en) 2013-11-29 2013-11-29 A kind of airborne sensor systematic error estimation based on cooperative target and compensation method

Country Status (1)

Country Link
CN (1) CN103616036B (en)

Families Citing this family (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104021292B (en) * 2014-06-06 2017-01-18 中国航空无线电电子研究所 Dim target detection and tracking method based on formation active networking
CN104849697B (en) * 2015-05-15 2017-11-14 上海前隆信息科技有限公司 It is a kind of based on the α β filtering methods for going inclined Coordinate Conversion
CN106202926B (en) * 2016-07-11 2018-08-14 河南大学 Space system deviation based on multi-node collaborative detection is registrated optimization method
CN106871927B (en) * 2017-01-05 2020-10-20 南京航空航天大学 Unmanned aerial vehicle photoelectric pod installation error calibration method
CN106969767B (en) * 2017-03-30 2020-03-17 西安交通大学 Estimation method for system deviation of moving platform sensor
CN107229037B (en) * 2017-06-01 2020-04-28 西南电子技术研究所(中国电子科技集团公司第十研究所) Mobile platform sensor measurement data dimension-expanding space registration method
CN107315171A (en) * 2017-07-02 2017-11-03 中国航空工业集团公司雷华电子技术研究所 A kind of radar network composite dbjective state and systematic error Combined estimator algorithm
CN108594193A (en) * 2018-04-24 2018-09-28 西安交通大学 A kind of radar system bias estimation method based on fixed target and noncooperative target
CN108919283A (en) * 2018-04-28 2018-11-30 北京空间飞行器总体设计部 Autonomous noncooperative target Relative Navigation and system on a kind of star
CN108716931A (en) * 2018-06-08 2018-10-30 中国航发沈阳发动机研究所 The disposition algorithm of aero-engine airborne sensor failure

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2742874A1 (en) * 1995-12-22 1997-06-27 Onera (Off Nat Aerospatiale) Method and system for detecting and locating an individual, for example developed under an avalanche deposit

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR2742874A1 (en) * 1995-12-22 1997-06-27 Onera (Off Nat Aerospatiale) Method and system for detecting and locating an individual, for example developed under an avalanche deposit

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
李力超;史伟.基于合作目标的三维空间贝叶斯传感器配准算法.《航空计算技术》.2013, *
杨波;王跃钢;彭辉煌.基于捷联惯导/里程计的车载高精度定位定向方法研究.《计算机测量与控制》.2011, *
熊伟;邢凤勇;潘旭东;彭应宁.基于合作目标的动平台传感器偏差估计方法.《系统工程与电子技术》.2011, *

Also Published As

Publication number Publication date
CN103616036A (en) 2014-03-05

Similar Documents

Publication Publication Date Title
Wu et al. Velocity/position integration formula part I: Application to in-flight coarse alignment
Groves et al. Intelligent urban positioning using multi-constellation GNSS with 3D mapping and NLOS signal detection
US9551561B2 (en) Determining location using magnetic fields from AC power lines
Obst et al. Multipath detection with 3D digital maps for robust multi-constellation GNSS/INS vehicle localization in urban areas
CN102829777B (en) Autonomous underwater vehicle combined navigation system and method
Skog et al. In-car positioning and navigation technologies—A survey
CN103528587B (en) Independent combined navigation system
St-Pierre et al. Comparison between the unscented Kalman filter and the extended Kalman filter for the position estimation module of an integrated navigation information system
US8515611B2 (en) Inertial measurement with an imaging sensor and a digitized map
Johannsson et al. Imaging sonar-aided navigation for autonomous underwater harbor surveillance
Xu et al. UWB-based indoor human localization with time-delayed data using EFIR filtering
US20170131407A1 (en) Selected aspects of advanced receiver autonomous integrity monitoring application to kalman filter based navigation filter
US6489922B1 (en) Passive/ranging/tracking processing method for collision avoidance guidance and control
Hsu Analysis and modeling GPS NLOS effect in highly urbanized area
CN104316903B (en) A kind of three station positioning using TDOA performance test appraisal procedures
Nordlund et al. Marginalized particle filter for accurate and reliable terrain-aided navigation
CN104729506B (en) A kind of unmanned plane Camera calibration method of visual information auxiliary
CN101520328B (en) Method for autonomous navigation using geomagnetic field line map
CN105043380A (en) Indoor navigation method based on a micro electro mechanical system, WiFi (Wireless Fidelity) positioning and magnetic field matching
CN104412065B (en) Positioner
RU2487419C1 (en) System for complex processing of information of radio navigation and self-contained navigation equipment for determining real values of aircraft navigation parameters
CN102096086B (en) Self-adaptive filtering method based on different measuring characteristics of GPS (Global Positioning System)/INS (Inertial Navigation System) integrated navigation system
CN104215259B (en) A kind of ins error bearing calibration based on earth magnetism modulus gradient and particle filter
ES2590761T3 (en) Issuer Geolocation
CN104406605B (en) Airborne many navigation sources integrated navigation analogue systems

Legal Events

Date Code Title Description
PB01 Publication
SE01 Entry into force of request for substantive examination
C10 Entry into substantive examination
GR01 Patent grant
C14 Grant of patent or utility model